The cellular regulatory protein p53 is mutated in up to one half of breast cancers. The alterations in p53 are generally single point mutations, which often result in over-expression of the mutant molecule in cancer cells. These mutations are not characteristically found in the regions of p53 bearing immunogenic peptides which bind to human (Hu) HLA A2.1. As a consequence of over-expression of mutant p53, the unaltered HLA A2.1 binding epitopes of p53 are also over-expressed, which makes them an attractive target for cellular immunotherapy. Unfortunately, previous investigators have failed to derive Hu p53-specific CTL which recognize full length p53 by in vitro immunization of Hu PBL. The proposed approach utilizes the generation of a xenogencic response against Hu p53 by a murine (Mu) transgenic model expressing the Hu HLA A2.l transplantation antigen. Mu CTL reactive against HLA A2.l binding peptides from Hu p53 will be derived by subcutaneous immunization along with adjuvant. The magnitude of the immune response will be measured against Hu p53 peptide primed or full length mutant p53 expressing Hu HLA A2.1+ target cells. Since the AA sequence of Mu and Hu p53 is conserved, it is likely that the highest affinity Mu CTL against p53 may be neonatally thymically deleted. To evaluate that hypothesis, CTL will also be derived from mice transgenic for HLA A2.l and knockout for Mu p53 (p53K0). Affinity of CTL derived from p53K0 mice for Hu mutant p53 expressing tumor cell lines will be compared to CTL derived from mice expressing Mu p53. p53K0 mice should not have been tolerized to p53, and should not have deleted high affinity p53 speciflc T cell receptors (TCR). Using PCR cloning techniques, the TCRs from high affinity T cell clones will be isolated. High affinity TCRs will be engineered into single chain TCR molecules, and introduced into naive T cells of the CD4 and CD8 subsets. The transduced T cells will be tested for their ability to kill HLA A2.l+, Hu p53 over-expressing, tumor cells in vitro and in in vivo tumor models. To address the possibility of generating autoimmunity, an autologous Mu p53 immunotherapy model wills-be generated. Mu p53 specific CTL will be derived, and the specificity for tumor as opposed to recognition of normal cells by molecularly cloned high affinity TCR transferred to naive Mu CTL will be tested in an in vivo Mu model. The ability to prevent carcinogen induced p53 over- expressing tumors will also be assessed.